AI-Enabled Fraud Detection System

An advanced machine learning-powered system for detecting fraudulent financial transactions. Features interactive dashboards, real-time analysis, and secure user authentication.

🌐 Live Demo

🔍 Features

• Machine Learning Detection: Detects suspicious activity using Random Forest and Isolation Forest algorithms
• Interactive Dashboard: Real-time statistics and visualization of transaction data
• CSV Upload & Analysis: Upload transaction data and analyze with custom sensitivity
• Comprehensive Reports: Generate detailed fraud analysis reports with visualizations
• Transaction Deep Dive: View risk factors and breakdowns for each transaction
• User Authentication: Secure registration, login, and session management

🛠️ Technology Stack

• Backend: Flask, SQLAlchemy, PostgreSQL
• Machine Learning: scikit-learn, pandas, NumPy
• Frontend: Bootstrap, Chart.js, JavaScript
• Authentication: Flask-Login

📋 Requirements

• Python 3.10+
• PostgreSQL database
• Modern web browser

🚀 Installation

1. Clone the repository

git clone https://github.com/Adithya17-star/AI-Enabled-Fraud-Detection.git
cd AI-Enabled-Fraud-Detection

2. Install dependencies

pip install -r requirements.txt

3. Configure environment variables

DATABASE_URL=postgresql://username:password@localhost:5432/fraud_detection
SESSION_SECRET=your-secure-secret-key

4. Initialize the database

python -c "from app import db; db.create_all()"

5. Run the application

python main.py

6. Open your browser and navigate to http://localhost:5000

📊 Usage Guide

Registration & Login

1. Create a new account via the Registration page
2. Log in using your credentials

Uploading Transactions

1. Navigate to the Upload page
2. Upload a CSV file with transaction data (see format below)
3. Preview the data and start analysis

Transaction CSV Format

Your CSV file should include the following columns:

• transaction_id: Unique identifier
• amount: Transaction amount
• timestamp: Date and time (YYYY-MM-DD HH:MM:SS)
• merchant: Merchant name
• merchant_category: Category of merchant (optional)
• description: Transaction description (optional)
• location: Transaction location (optional)
• ip_address: IP address (optional)
• device_id: Device identifier (optional)

Analyzing Transactions

1. Select detection model (Random Forest or Isolation Forest)
2. Adjust detection sensitivity
3. Start analysis to identify fraudulent transactions

Dashboard

• View key metrics including total transactions, fraud count, and amounts
• Analyze transaction trends over time
• Examine fraud distribution by merchant category

Reports

1. Create a new report by specifying date range and title
2. View detailed fraud patterns and statistics
3. Export reports for record-keeping

📝 Project Structure

AI-Enabled-Fraud-Detection/
├── static/                     # Static assets
│   ├── css/                    # CSS stylesheets
│   └── js/                     # JavaScript files
├── templates/                  # HTML templates
├── app.py                      # Flask application setup
├── main.py                     # Application entry point
├── models.py                   # Database models
├── routes.py                   # Application routes
├── fraud_detection.py          # Fraud detection algorithms
├── data_preprocessing.py       # Data preprocessing utilities
└── utils.py                    # Helper functions

🔒 Security Considerations

• All passwords are securely hashed using werkzeug's security functions
• PostgreSQL database with secure connection
• Input validation on all form submissions
• Session management via Flask-Login

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

👥 Authors

• Adithya Kakarla - Initial work - Adithya17-star

🙏 Acknowledgements

• Bootstrap for the UI components
• Chart.js for data visualization
• Flask for the web framework
• scikit-learn for machine learning capabilities

---

This project is part of an ongoing effort to improve fraud detection systems using artificial intelligence and machine learning techniques.